The proposed studies are concerned with critically evaluating the effectiveness of a variety of anti-endotoxemic agents, with special emphasis on hyperimmune sera and massive doses of adrenal corticosteroids, in reducing mortality during antibiotic-treated experimental gram-negative bacterial sepsis. Models of gram-negative sepsis have been developed in Swiss-Webster mice, employing a variety of bacterial species inoculated with I.P. and I.V. LD85-95 dosages. When appropriate antibiotic therapy is administered concomitantly with the bacterial challenges, no mortality results. When antibiotic therapy is delayed, mortality increases progressively. The delay time that yields an LD50 has been established for each infection. At this point, treatment with the anti-endotoxemic agent will be initiated. Antisera to the homologous smooth infecting organisms and to a variety of rough gram-negative bacterial mutants (including the Re 595 mutant of S. minnesota and the J5 E. coli 0111 mutant) will be administered I.V., concomitantly with normal sera for control purposes. In other studies, methylprednisolone sodium succinate (30 mg/kg) or dexamethasone sodium phosphate (8 mg/kg) will be adminstered I.V. or I.M. Other anti-endotoxemic agents to be evaluated include anticoagulants, anti-proteolytic substances (Trasylol), salicylates and glucose. Both single and repetitive injections will be employed. The above studies will be performed in both previously healthy animals and in those whose resistance to infection has been impaired by alkylating agents (cyclophosphamide). The major aim of these studies will be to evaluate the efficacy of hyperimmune sera and of adrenal corticosteroids, and subsequently of other anti-endotoxemic agents, in reducing mortality from gram-negative bacterial sepsis when administered after sepsis has been initiated and at a time when appropriate antibiotic therapy alone no longer is capable of eliminating mortality.